Insight-Driven Retailing Blog

I downloaded two more iPhone applications from retailers and played with them a bit. The first is Gucci's app, which is unique presumably because their customers are unique. They run "iPhone only" specials like $500 designers sneakers (limit one per customer, please). The've attempted to capture the tastes of their designers by displaying their music playlists and places they visit. I suppose this makes sense for their audience, but it didn't do much for me. The interesting feature I found was the ability to post products to Facebook to share with friends.

Walmart's iPhone application has a similar feature. It allows users to post a product to Facebook so that you can ask for advice. I tried this as you can see in the picture. Seems like a pretty handy feature. It also allows users to "bookmark" products, sort of like a wishlist although it doesn't appear to be linked to their e-commerce site. Bummer.

The Walmart application is limited to electronics, which is a good place for them to start. There's an innovative feature that lets you take a picture of your room and determine the best TV size. I didn't bother to really see if it works well (I took a picture of the first class cabin on a 757. It said it needed a 20" TV, which is actually about right.), but it demonstrates outside-the-box thinking.

Any other retail-industry apps I should be using?

If you've followed the hype surrounding Wolfram Alpha, you know that search is slowly changing. Typically, people use search to find information that is then used to make an informed decision. That's traditionally how it works with product searches as well. For instance, searching for a digital camera yields several different results, including sellers, reviews, and shopping aggregators. Now you are left to comb through the different sites, noting information in order to make a purchase decision.

But here's the problem. That's actually lots of work. Some of the prices include tax and shipping, but some don't. One or two reviews don't actually constitute a trustworthy assessment. You need to dig up lots of perspectives and look for trends. And you should probably search the coupon sites as well so you're not passing up an even better deal.

So while the Web and search engines are great at finding tons of information, they don't do much to help process the information into something usable. That's where search is starting to change. Microsoft's Product xRank (from Bing) has attempted to better aggregate information into usable data. For example, their product rankings help aggregate information and shortcut the decision process.

We're closing in on the ability to ask, "Where can I buy the best compact, point-and-shoot digital camera for the lowest price?" Give it a couple more years.

A year ago Larry Ellison announced Exadata, Oracle's latest attempt to get into the hardware business. It was followed by the announcement of Exadata v2 (video), which significantly improves upon the first version. I don't think that Oracle wants to expand into hardware per se, but instead is targeting systems. The difference is optimization. When hardware and software are optimized to work together, you can offer customers a solution that not only simplifies implementations, but also offers greater functionality and performance. Enter Exadata.

Strictly speaking, Exadata is the intelligent storage server, not the entire solution. The Database Machine is the combination of multiple Exadata storage servers, database servers, and network switches pre-configured and hooked together in a rack. The Database Machine is the whole system, but its frequently referred to as "Exadata" since that's a more unique brand.

Typically applications "ask" the database a question in the form of a SQL query, like "who are all my customers in Texas?" To answer this question, the database server asks the storage for all the data in a table, then does comparisons to find the matching records. There are two big benefits of using Exadata storage servers. First, they are intelligent so they will do the comparisons and return only matching records. That reduces the load on the database server and the network that connects them. Second, Exadata storage servers use flash technology, which is faster than disks and cheaper than memory.

The Database Machine simplifies implementations because the storage, database servers, and networking are all included and optimized to work together. It uses Oracle Enterprise Linux and the Oracle 11gR2 database with RAC. Its also fully redundant and scalable. No big decisions, no expensive consultants, no arguments over configurations.

Oracle Retail plans to benchmark both our data warehouse and transactional applications on Exadata and measure the performance gains. The retail industry is in desperate need of ways to increase performance while reducing costs, and I expect this "systems" approach will yield awesome results. Stay tuned.

I was very saddened to hear that David Taylor, founder of the PCI Knowledge Base, unexpectedly died of a heart-attack yesterday. Apparently his website will live on. Evan Shuman's note says it better than I could.

There are three related yet often confused terms that creep into conversations about retail. I started a conversation on that topic over at the ARTS Linked-in discussion board that has some interesting comments. The three terms are (with definitions from Wikipedia):

Hierarchy "is an arrangement of items (objects, names, values, categories, etc.) in which the items are represented as being "above," "below," or "at the same level as" one another and with only one "neighbor" above and below each level."

Taxonomy "is the practice and science of classification. Typically this is organized by supertype-subtype relationships, also called generalization-specialization relationships, or less formally, parent-child relationships."

Ontology "is a formal representation of a set of concepts within a domain and the relationships between those concepts. It is used to reason about the properties of that domain, and may be used to define the domain."

Did I loose you on that last one? Ignore it for now. When we classify merchandise, do we use a hierarchy or taxonomy? The generally accepted term in the retail industry is "merchandise hierarchy" but I'm not so sure its accurate.

The classic example of a hierarchy is that of the organization. The lieutenant reports to the captain who reports to the major in the Army hierarchy. For a taxonomy, the classic example is the classification of animals. The dog is a canine, carnivore, mammal (sparing you lots of Greek and Latin words).

So which term fits Electronics --> Televisions --> Plasma --> Panasonic TH-50PX60U best? Looks like a taxonomy to me because each successive layer gets more detailed. I'm amazed at the lack of standards in this area. Supplier/retailer data pools should be much simpler than they are.

Which brings me to my final rant. Until we, as an industry, straighten this out we'll never achieve the machine-to-machine communications we strive for in the Semantic Web. This is where the term "ontology" becomes relevant, as it provides the discipline to define a constant view of the retail domain that moves us from information to answers.

RetailWire ran articles yesterday and today on the success of iPhone applications for the retail industry. Apps from Amazon, Target, Gap, and Whole Foods seem to be getting traction, but all of them had shortcomings. Here's my list for what makes an iPhone app successful:

1. Utility -- the application has to be useful,usually by saving me time. If I can do something easier on my PC, I will so don't duplicate it on an iPhone. A big part of this is taking advantage of mobility and the iPhone's unique features, like location awareness. For example, the Dunkin Donuts application allows a person to collect orders from friends before going to Duncan Donuts.
2. Intuitive -- if I have to guess at anything, I quit. I've been spoiled by iPhone user interface and expect all applications to be very easy to use. The Amazon application is easy to use, and the "remembers" functionality that allows you take pictures of things to remember is easy to use.
3. Entertaining -- if the application is amusing, I tend to let friends know about it, but it doesn't stay on my iPhone for very long. This is great for applications that are focused on short-term marketing programs. For example, I recently downloaded the Amp Energy (drink) application only because its funny (and politically incorrect). Allowing social features helps to keep the content fresh.

Most applications meet 2 out of 3, which is just enough to keep them on my iPhone. But even that standard doesn't ensure success. The best applications will lie unused if no one knows about them, so a certain amount of marketing is necessary. For example, both Sherwin Williams and Benjamin Moore created iPhone applications that allow the user to take a picture, then select a color within the picture and be told the matching paint color. Even though Sherwin released their application first (May 12), Moore (June 1) had better marketing and therefore more downloads. Here's a quick video from AdAge that talks about the Ben ColorCapture application.

As described in last week's posting, RedLaser is an iPhone application that accurately scans product barcodes using the built-in camera. The product was created by University of Michigan graduates Jeff Powers and Vikas Reddy through a start-up called Occipital. I was fortunate enough to be able to interview co-founder Jeff Powers, who left U of M to commercialize on his computer vision interests.

Why did you decide to create RedLaser?

We realized, despite 50,000 apps at the time, there was no barcode scanner for the iPhone that actually worked. There were a few that required an add-on lens, but nothing that could do it well without the lens. There was clearly a need for such a thing, with quite a bit of chatter around the web incited by the fact that Android phones had a decent ability to scan barcodes due to their improved camera. As a boostrapped company, we saw RedLaser as a clear opportunity to keep the company afloat, as long as we could actually execute and create something that worked.

What separates RedLaser from similar iPhone apps?

The barcode recognition algorithms in RedLaser are state of the art. We've explicitly compensated for blur, lighting, curvature, among other parameters, making RedLaser extremely robust. We also went as deep as ARM assembly to make it work fast, making full use of the available CPU. Similar apps mostly use open source barcode software, which performs miserably on low quality images. There's only one other application that attempts to deal with poor quality images, but it doesn't perform as well as RedLaser.

How do you think RedLaser will change the way people shop?

Retailers enjoy being able to control what information consumers have at the point of purchase. But with RedLaser, people can take back some control and immediately know if it's a good price, or link through to find out if others think it's a good product. In the short term, RedLaser is already helping people get better deals from retailers that have price match guarantees..

Where do you see this all going?

Right now, it's mostly about online retailers getting increased visibility. Longer-term, I think we'll see physical retailers making their items visible online as well. This will help physical retailers compete with online, and give consumers even more choices and information.

Technology-wise, barcodes are just one way we can use mobile cameras to access information about products and real-world objects. Occipital is expandng mobile visual processing capabilities, which should lead to exciting new applications in the near future.

Frequent readers know I like to experiment with iPhone applications as I believe mobile commerce will continue to gain even more acceptance. In past reviews, I've been disappointed with iPhone barcode scanners but always assumed the problem was a lack of resolution on my 3G. Recently I downloaded and tried a new iPhone application called RedLaser that seems to work much, much better even on my 3G.

The RedLaser application is dirt-simple. When the camera activates, just aim the cross-hair at a barcode and when the software is happy, it automatically takes the picture. No need to press a button and wonder if its a good shot. I tried this in two stores (Fry's and Toys-R-Us) under different lighting conditions and it worked each time.

Once the barcode has been scanned, the application does product searches on Amazon and Google to find prices. For example, while in Fry's there was a Buffalo DriveStation (external harddrive) for $86. Is that a good deal? I used RedLaser to find that other online stores were selling it for $100-132 so the $86 price was a bargain. Yes, I could have manually entered the barcode but its so much easier to just snap a picture.

And RedLaser is available to other iPhone application developers as software development kit, so I expect new, innovative uses to follow.

As I described last year, Oracle Retail POS ran the bookstore at OpenWorld again this year. It's a great opportunity to see our software in action, as Jeff Grossman describes in this video.

For some reason, this has been the year for POS. The "perfect storm" has been brewing for a while, but we're only now seeing a big uptick in interest and implementations. Asking around for the reason, I've heard answers such as "retailers are limiting their investments to things that directly touch the customer," and "in order to feed the backoffice systems, retailers need better and faster sales data." I think another reason might be that retailers want to be ready when the economy rebounds, so taking the time to replace aging systems now is a strategic investment.

The theme of "Complete, Open, Integrated" continued this year, as described by Duncan Angove in this interview that was shot following his retail keynote. He goes on to explain why middleware matters and the advantage it brings to Oracle. Duncan predicts that retailers will "jump the curve" by combining facets of traditional merchandising with those from customer-centric software, like Siebel. Using Web 2.0 concepts to better understand customer demand will become a competitive advantage, and retailers need a strategy in this regard sooner rather than later.

As Joe Skorupa pointed out in his blog posting on OpenWorld, retail has finally received the spotlight. He's referring to Duncan Angove's presentation during the general session on Monday. Duncan explained how Oracle Retail used middleware to integrate existing products specifically for fashion retailers to support "fast fashion." While the Oracle Retail Integrated Fashion Planning solution has been around for a few months, the Oracle Retail Integrated Inventory Planning solution was just announced today. Both of these solutions have the same key ingredients: business process, analytics, and user experience.

Business Process
Oracle's Retail Reference Model is a set of business process models that capture best practice processes validated by customers, analysts, and partners. It illustrates, at varying levels of detail, the automated and manual steps involved for item induction, assortment execution, replenishment, etc. They show which roles are involved, how applications interact, and where alerts occurs.

Analytics
In retail, there are two sides to analytics. First, retailers must capture and process data so they can detect trends and measure key performance indicators. This is necessary to establish baselines and continuously build better plans. You know this as "business intelligence." Second, science in the form of statistical algorithms are applied to optimize the decisions made, such as promotions, markdowns, and replenishment.

User Experience
No dinner is complete without the ambiance. The business processes are served up with pervasive analytics in a compelling Web 2.0 UI that includes role-based dashboards, application navigation, and collaboration. This is done via our retail specific extensions to Oracle WebCenter which we have branded Oracle Retail Workspace.

I love using the xobni plug-in for Outlook because as I read emails, xobni shows me information about the sender, a history of their conversations and attachments, and even provides me analytics about their email behavior. This information is provided in-context with no additional effort on my part and allows me to more quickly respond to emails. We're going for the same experience with Workspace, hoping to increase productivity for the retail industry. The Integrated Fashion Planning and Integrated Inventory Planning solutions are just the start.